The Biochemical Mechanisms of Resistance by ... - ACS Publications

produces normal levels of the drug-sensitive enzymes and differs from the parent strain in its ability to con- centrate alanine. Since D-cycloserine i...
0 downloads 0 Views 950KB Size
VOL.

6,

NO.

8,

AUGUST

1967

The Biochemical Mechanisms of Resistance by Streptococci to the Antibiotics D-Cycloserine and 0-Carbamyl+-serine" Richard H. Reitz,t Hutton D. Slade,f and Francis C. Neuhaus!

has elevated levels of D-alanine :D-alanine ligase (ADP) (five times) and alanine racemase (eight times) and does not differ from the parent strain in its ability to concentrate alanine. Alanine racemase and D-alanine : D-alanine ligase (ADP) in mutant @-CS,) are similar to the enzymes in the parent strain with respect to Michaelis-Menten constants, susceptibility to inhibition by D-cycloserine, and heat stability. The second type of D-cycloserine resistant mutant, mutant @-csb), produces normal levels of the drug-sensitive enzymes and differs from the parent strain in its ability to concentrate alanine. Since D-cycloserine inhibits the accumulation of alanine by the parent strain and since mutant @-csb) possesses a defective transport system for alanine, it would appear that this mutant is unable to concentrate D-cycloserine. Transformation experiments with deoxyribonucleic acid prepared from mutant (D-CSJ suggest the existence of two genetic markers associated with resistance to D-cycloserine in this strain. One marker is associated with increased levels of alanine racemase whereas the other marker is associated with increased levels of D-alanine :D-alanine ligase (ADP).

The antibacterial activity of D-cycloserine (~-4-amino-3-isoxazolidone) (D-CS) and O-carbamylD-serine (0-CS) has been correlated with the inhibition of enzymes necessary for peptidoglycan biosynthesis. D-Cycloserine inhibits alanine racemase (EC 5.1.1.1) and D-alanine malanine ligase (ADP) (EC 6.3.2.4) (Strominger, J. L., Ito, E., and Threnn, R. H. (1960), J. Am. Chem. SOC.82, 998). 0-Carbamyl-D-serine inhibits only alanine racemase (Lynch, J. L., and Neuhaus, F. C. (1966), J. Bacteriol. 91, 449). Mutants of Streptococcus strain Challis selected for resistance to 0carbamyl-D-serine have increased levels of alanine racemase. Enzymes that are not sensitive to O-carbamylD-serine @-alanine :D-alanine ligase (ADP), UDP-NAcmuramyl-L-Ala-D-isoGlu-L-Lys :D-Ala-D-Ala ligase (ADP) (EC 6.3.2.10): UDP-NAc-muramyl-L-Ala-DisoG1u:L-lysine ligase (ADP) (EC 6.3.2.7)) are not elevated in mutant (0-CS). Resistance to O-carbamylD-serine develops in steps, and each increase in resistance is accompanied by an increase in the specific activity of alanine racemase. In the most resistant mutant, alanine racemase is increased eightfold. Two types of D-cycloserine-resistant mutants have been isolated and examined. The first type, mutant @-CS,), ABSTRACT:

C

Ciak and Hahn, 1959; Tanaka et al., 1963). Cultures of bacteria that have been grown in the presence of these antibiotics have defective walls and are susceptible to lysis (Shockman, 1959; Ciak and Hahn, 1959). With each antibiotic there is an accumulation of the incomplete cell wall precursor, UDP1-NAc-muramylL-Ala-D-isoGlu-L-Lys, which lacks the terminal D-Ala-D-Ala moiety (Strominger et al., 1959; Lynch and Neuhaus, 1966). The addition of D-alanine to the medium overcomes the inhibition of growth by D-cycloserine and 0-carbamyl-D-serine and decreases the accumulation of UDP-NAc-muramyl-L-Ala-DisoGlu-L-Lys (Strominger et al., 1959; Lynch and Neuhaus, 1966). These results have been correlated with the effects of the analogs on enzymes which catalyze the incorporation of D-alanine from L-alanine into the wall precursor. D-Cycloserine is an effective inhibitor of D-alanine :D-alanine ligase (EC 6.3.2.4)

ycloserine (~-4-amino-3-isoxazolidone)and 0carbamyh-serine, analogs of D-alanine, are inhibitors of peptidoglycan biosynthesis (Park, 1958, 1960; D-

* From the Departments of Chemistry (Biochemistry Division) and Microbiology, Northwestern University, Evanston and Chicago, Illinois. Received February 20, 1967. Supported in part by a grant (AI-04615) from the National Institute of Allergy and Infectious Diseases (F. C. N.); by a grant-in-aid (F. C. N.) from Eli-Lilly and Co., Indianapolis, Ind.; by a Public Health Service training grant ( 5 TI GM-626); by Grants HE-11116 (formerly GM- 10006) (F. C. N.) and HE-03709 (H. D. S.) from the National Heart Institute; and by grants from the Chicago, Illinois Heart Association (H. D. S.) and Office of Naval Research (H. D. S.). Taken in part from a thesis submitted by R. H. R. in partial fulfillment of the requirements for the Ph.D. degree from Northwestern University. A preliminary report of this work has been presented (Reitz et a!., 1966). t Supported in part by a Public Health Service Fellowship (Fl-GM-23597) from the National Institute of General Medical Sciences. Present address: Dow Chemical Co., Midland, Mich. $ Research Career Awardee K6-GM-16284, National Institute of General Medical Sciences, National Institutes of Health. $Supported by U. S. Public Health Service Research Career Development Program Award 1-K3-AI-6950 from the National Institute of Allergy and Infectious Diseases. To whom reprint requests should be sent.

RESISTANCE

1 Abbreviations used: UDP, uridine diphosphate; ADP, adenosine diphosphate; ATP, adenosine triphosphate; FUDP, 5-fluorouridine S'diphosphate; D-CS, D-4-amino-3-isoxazolidone; 0-CS, 0-carbamyl-u-serine.

TO

D-CYCLOSERINE

AND

2561

0-CARBAMYL-D-SERI

BIOCHEMISTKY

2562

(ADP) (D-Ala-D-Ala synthetase) (Strominger et d., were prepared immediately before use. In every case, 1960; Neuhaus and Lynch, 1964) and alanine racemase D-cycloserine and 0-carbamyl-D-serine were sterilized (EC 5.1.1.1) (Strominger et al., 1960). In contrast, by passage through a Millipore filter (0.2 p). Colonies 0-carbamyl-D-serine inhibits only alanine racemase from the plates with the highest concentration of anti(Lynch and Neuhaus, 1966). D-Cycloserine and 0biotic were transferred to Todd-Hewitt broth. This carbamyl-D-serine do not inhibit the enzyme that adds cycle was repeated until a high degree of resistance was D-Ala-D-Ala to UDP-NAc-muramyl-~-Ala-~-isoGluobtained. The antibiotic sensitivity was quantitated in a medium which contained 1 % peptone T, 1 % L - L ~(Strominger, s 1962; Lynch and Neuhaus, 1966). The emergence of resistance to D-cycloserine has yeast extract, 0.5% K2HP04, and 0.5% glucose. The minimal inhibitory concentration is defined as the been observed in many clinical situations (for a review, lowest concentration of antibiotic that limits the culsee Neuhaus, 1967). In addition, D-cycloserine-resistant ture to a turbidity of 0.9 under conditions in mutants have been isolated in the laboratory (Howe which the control tube reaches 1.3. Turbidimetric et al., 1964; Chambers et al., 1963; Curtiss et al., 1965). measurements were performed in a Bausch and Lomb Curtiss et ai. (1965) studied the kinetic and genetic Spectronic 20 in 18 X 150 mm test tubes at 650 mp. aspects of D-cycloserine resistance in Escherichia coli. The mutant strains resemble the parent strain in cell It is the purpose of this paper to establish the biomorphology, Gram stain, colonial appearance, cichemical mechanism(s) of resistance to D-cycloserine hemolysis on blood agar, and group H typing reaction and 0-carbamyl+-serine in Streptococcus (serological (Swift et al., 1943). The strains were maintained as group H) strain Challis. lyophilized cultures in cacuo at -20". The mutants that have been studied in detail are summarized in Experimental Section Table I. Mutant (D-CS,) has been in culture 5 years Muteriais ~-[l-'~C]Alanine,L-[ l-14C]alanine, and L-[U-'~C]lysine were purchased from California Corp. for Bio'TABLE I: Summary of Mutants Used. chemical Research. The plastic beads (styrenedivinylbenzyne copolymer, 20-50 mesh, 8 % cross-linked), Minimal Inhibitory ConcenDowex 50-X8 (100-200 mesh), and Dowex 1 were tration (moles/l.) gifts of the Dow Chemical Co. We are indebted to Dr. 0. K. Behrens of the Eli Lilly Co., and Drs. M. C. Streptococcus strain 0-CarbamylBachman and P. H. Hidy of Commercial Solvents Challis D-Cycloserine D-serine Corp. for D-cycloserine and 0-carbamyb-serine, respectively. Todd-Hewitt broth, blood agar base, Parent 5 x 10-4 2 x 10-3 proteose peptone, and yeast extract were obtained Mutant (D-CS,) 100 x 10-4 30 x 10-3 from Difco Co. Defibrinated sheep's blood was supMutant (D-CSb) 40 x 10-4 b plied by Dr. Robert Galvin (Beecher, Ill.). D-Amino Mutant (0-CS). 20 x io-' 20 x 10-3 acid oxidase (EC 1.4.3.3) (4.5 units/mg, electrophoreti= See Table VI for summary of isolation. b Not cally purified) and catalase (CTR) were purchased determined. from Worthington Biochemical Corp. For the D-AlaD-Ala adding enzyme, a 55-70% ammonium sulfate fraction (0.12 unit/mg) was used (Neuhaus and Struve, 1965). and upon retesting has demonstrated a stable mutation D-[14C]Ala-~-[ 4C]kla and UDP-NAc-muramyl-Lfor D-cycloserine resistance. Ala-D-isoGlu-L-Lys were prepared according to the procedures described by Neuhaus and Struve (1965) Preparation of Cell-Free Extracts. The bacteria were andstickgold and Neuhaus (1967). UDP-NAc-muramylgrown at 37" in 500 ml of a medium which contained L-Ala-D-isoGlu was prepared with Staphylococcus 1 % peptone T, 1 % yeast extract, 0.5% K2HP04, ciureus Copenhagen in a manner similar to that described and 0.5% glucose. When the turbidity had reached by Strominger and Threnn (1959) (see Reitz, 1966). 0.7, the cells were chilled and harvested by centrifugaThe isolation of the nucleotide from the extract is tion at 6000g for 20 min. The bacteria (1.6-2.2 g wet identical with that described by Stickgold and Neuhaus wt) were washed once in cold 0.01 M Tris-HC1 buffer (1967) for FUDP-NAc-muramyl-pentapeptide. Cell (pH 7.8) and then resuspended in 20 ml of the same walls were prepared according to the method described buffer containing 17.5 g of plastic beads and 2 drops by Struve et ai. (1966). of antifoam. The bacteria were disrupted in a Bronwill Mutant Isolation. Streptococcus strain Challis (107mechanical cell homogenizer (Braun Model MSK) lo8 cells) from an 18-hr culture (Todd-Hewitt broth) at 4000 cycles/min for 5 min. During the disruption was spread on a blood agar plate containing 4% blood the temperature was maintained between 2 and 8" agar base, 4 % defibrinated sheep's blood, and either with liquid carbon dioxide. After removing the beads D-cycloserine or 0-carbamyl-D-serine (e.g., see Table by filtration through a 110-mesh nylon cloth, the VI). Since D-cycloserine slowly loses its bactericidal unbroken cells and cell walls were removed by centrifugation at 14,OOOg for 10 min. The extracts were activity in agar plates (Curtiss et al., 1965), the plates

R E I T Z , SL.ADF., A N D N F U H A L I S

VOL.

6,

NO.

8,

A U G U S T

1967

adding enzyme (Neuhaus and Struve, 1965). Under frozen in 0.6-ml polyethylene capsules and stored in liquid nitrogen until used. In order to establish the conditions of the second stage, 0.2 pmole of D[1c]Ala-~-[14C]Alawas quantitatively transferred to meaningful enzyme profiles, each capsule was thawed the nucleotide tripeptide. The amount of UDP-NAconly once. muramyl-L-Ala-D-isoGlu-L-Lys-~-[ 14C]Ala-~-[14C]Ala Assays. The unit used in each assay is defined as 1 was proportional to the concentration of D-Ala-n-Ala pmole of product formed/hr. The D-Ala-D-Ala adding synthetase added. enzyme was assayed according to the method described Alanine Accuniulution Assay. The amount of [ 14C]by Neuhaus and Struve (1965). Alanine racemase was alanine concentrated by cells of Streptococcus strain assayed in the %-alanine to D-alanine assay" described Challis (or mutants) was established in the following by Lynch and Neuhaus (1966). procedure. The bacteria were grown in 500 ml of L-Lysine Adding Assay. The L-lysine adding enzyme medium that contained 1% peptone T, 1 % yeast exwas assayed by the addition of ~ - [ ~ ~ C ] l y sto i n eUDPtract, 0.5% K2HP04, and 0.5% glucose. When the NAc-muramyl-L-Ala-D-isoGluwith the formation turbidity reached 0.8, the cultures were chilled (4") of UDP-NAc-muramyl-~-Ala-~-isoGlu-~-[~~C]Lys. The and harvested by centrifugation at 6000g for 20 min. standard assay mixture contained 0.4 pmole of ATP The cells (2 g) were resuspended in 100 ml of 0.05 M neutralized with NaOH, 0.4 pmole of MgCI2, 10 pmoles potassium phosphate buffer (pH 8.0) and recentrifuged. of Tris-HC1 bulfer (pH 7.8), 32 nmoles of UDP-NAcThe pellet of cells was suspended in 20 ml of the same muramyl-L-Ala-D-isoGlu, 0.1 pmole of L-[ 14C]lysine buffer and equilibrated at 25 '. (6 X l o 4 cpm), and enzyme preparation in a The cells were preloaded with alanine in a procedure total volume of 0.10 ml. The composition of the similar to that described by Winkler and Wilson mixture is modified from that described by Ito and (1966). The preloading was accomplished in the followStrominger (1962b). The sample of extract was added to ing incubation mixture: 180 mg of cells (wet weight), the assay mixture at 37 " and incubated for 10 rnin at this 20 mg of glucose, 2 pmoles of D- or L-alanine, and temperature. The reaction was terminated by the addi90 pmoles of potassium phosphate bufer (pH 8.0) tion of 0.5 ml of 0.2 N sodium citrate buffer (pH 2.2). The separation of UDP-NAc-muramyl-L-Ala-D-isoGlu- in a volume of 2 ml. The cells were incubated for 5 min at 25". After preloading, 50 nmoles of [I4C]~ - [ l ~ C ] L yfrom s ~-[1~C]lysine is identical with that dealanine ( 5 x l o 5 cpm) was added and the cells were scribed for the D-Ala-D-Ala Adding Assay (Neuhaus allowed to accumulate the [14C]amino acid for the and Struve, 1965). With these assay conditions, the desired length of time. The uptake was terminated amount of product is proportional to the enzyme conby adding 4.0 ml of 0.05 M potassium phosphate buffer centration. (pH 8.0) (0') and immediately immersing each tube D - A l a - D - A h Synthetase Assay. Previously, the routine in an ice-water bath. The tubes were immediately assay for D-Ala-D-Ala synthetase in extracts involved the separation of D-[14C]alanine from D - [ I ~ C ] A ~ ~ - D - centrifuged at 27,000~for 5 rnin at 0-4". The cells were then washed with 5 ml of 0.05 M phosphate ["ClAla by paper chromatography (Neuhaus, 1962). buffer (0-4") (pH 8.0). After centrifuging, the pellet The P, Assay and ADP Assay that were described was resuspended in 1 ml of distilled water and transare not satisfactory for the assay of this enzyme in ferred to a vial containing 10 ml of toluene-Triton extracts. In the present work this enzyme was assayed X-100 scintillation fluid (Patterson and Greene, 1965). by adding the product of the enzyme reaction, D[14C]Ala-~-[14C]Ala, to UDP-NAc-muramyl-L-Ala-DTransformation of D-Cycloserine Resistance. For the transformation of D-cycloserine resistance, DNA isoGlu-L-Lys and separating the product from the remaining ~-['~C]alanine in the incubation. was prepared from mutant (D-CS,) by the method of Perry and Slade (1962). Transformation was perThe reaction mixture contained 5 pmoles of Trisformed according to the procedure described by Perry HC1 buffer (pH 7.8), 2.5 pmoles of KCl, 0.5 pmole of and Slade (1966). After 15 rnin DNase (10 pg) was MgC12, 0.5 pmole of ATP neutralized with NaOH, added to terminate the transformation. The cells were 2 pmoles of ~ - [ ~ ~ C ] a l a n i(25,000 ne cpm/pmole), and then allowed to incubate an additional 2 hr at 37" enzyme preparation in a total volume of 0.05 ml. before plating samples on blood agar plates containing The enzyme was added to the reaction mixture at 37" 300 pg of D-cycloserine!ml. After 48 hr at 37", the and incubated for 10 rnin a t this temperature. The plates were examined for transformants. About 12 reaction was terminated by placing the tube in a boiling of the viable cells exposed to DNA was able to form water bath for 2 min. colonies on the plates containing 300 pg/ml of D - C ~ C ~ O In the second stage, a solution (0.05 ml) containing 0.25 pmole of UDP-NAc-muramyl-L-Ala-D-isoGlu-L- serine. When a control culture of the Streptococcus strain Challis (DNase added before the transforming Lys, 0.20 pmole of ATP neutralized with NaOH, 2 DNA) was carried through this procedure and plated ymoles of MgCl?, and 500 pg of D-Ala-D-Ala adding (105 cells/plate), no D-cycloserine-resistant colonies enzyme was added, and the mixture was incubated were observed. for 20 min at 37". The reaction was terminated by the Analytical Procedures. Protein was determined by addition of 0.5 ml of 0.2 N sodium citrate buffer (pH 2.2). The UDP-NAc-muramyl-L-Ala-D-isoGlu-L-Lys-the method of Lowry et al. (1951). For the analysis D-[ 14C]f%la-~-[14C]Ala was separated from ~ - [ ~ ~ C ] a l a - of amino acids, the walls (1.7 mg wet wt) were hydrolyzed in 5.7 N HC1 for 12 hr and the amino acids were nine by the procedure described for the D-Ala-D-Ala

RFFISTANCF TO

T)-CYCI OSFRINF

2563

AND O-Ci\RR4MYI.-D-FFRINF

BIOCHEMISTRY

determined on the Beckman Model 120 amino acid analyzer according to the metllods described by Spa&man et al. (1958). Measurements of radioactivity were made in polyethylene vials using the Packard Tri-Carb liquid scintillation spectrometer (Model 314-EX). The scintillation fluid (15 ml) was described by Patterson and Greene (1965) and evaluated by Benson (1966).

Specific Activities as a Function of the Growth Phase in Streptococci4s Strain Challis and MutantD-CSa.a TABLE 111:

- .-

__ Specific Activity (units/mg of protein) 0.4

Enzyme

Results Resistance to D-Cycloserine. Cultures of Streptococcus strain Challis, Streptococcus strain Challis @-CS,), and Streptococcus strain Challis @-CS,) were grown under identical conditions to a turbidity of 0.7 and harvested. Cell-free extracts were prepared from these cultures and assayed for the following enzymes: (1) alanine racemase, (2) D-Ala-D-Ala synthetase, (3) D-Ala-D-Ala adding enzyme, and (4) L-lysine adding enzyme. The specific activities of these enzymes are summarized in Table 11. In the case of mutant @-CS,),

TABLE 11: Enzyme Profiles of Streptococcus Strain Challis, Mutant D-CS,, and Mutant D-CSb:

Specific Activity (units/mg of protein)

Alanine racemase D-Ala-D-Akt synthetase D-Ala-D-Ala adding L-Lysine adding

-

1.2

* 0.1

9.6 f 1.0

1.7

0.033 f 0.003 0.16 f 0.02 0.043 0.35 i 0.15

0.29

=I=

0.15

-

0.12 i 0.01

0.17

=I=

0.01

-

.The results for the parent and mutant (D-CSa) are the average of five separate experiments. Blank spaces not determined.

Parent Alanine racemase D-Ala-D-Ala synthetase D-Ala-D-Ala adding

Turbidity 0.7

1.0

1.3 1.2 1.1 0.030 0.033 0.031 0.26 0.42 0.46

Mutant D-CS, Alanine racemase D-Ala-D-Ala synthetase D-Ala-D-Akt adding

10.7 0.16 0.30

9.6 0.16 0.28

8.3 0.17 0.35

.The assays and growth conditions are described in the Experimental Section.

D-Ala-D-Ala synthetase are independent of the growth phase. The increase in specific activities observed with mutant @-CS,) could be either the result of an endogenous inhibitor in the parent strain or the result of a limitation in an essential cofactor. If either of these were correct, one would observe an apparent stimulation or inhibition of the activity in one extract by the addition of the other extract. The results in Table IV illustrate that the sum of the activities in the extracts assayed separately equals the activities when the extracts are assayed simultaneously. Thus, the increased specific activities do not appear to be the result of an endoge-

TABLE IV: Summation of Activities in Extracts from the Parent and Mutant (D-CS,).

Activity (nmoles/lO min)

2564

an eight- and fivefold increase was observed for alanine racemase and D-Ala-D-Ala synthetase, respectively. The specific activities of the L-lysine adding enzyme and the D-Ala-D-Ala adding enzyme are not significantly different from that observed for the parent strain. With mutant (D-CSb) no significant increase in the specific activities of either alanine racemase or DAla-D-Ala synthetase was detected. Since the specific activities of the enzymes may be a function of the growth phase (e.g., see Ito and Strominger, 1962a), extracts were prepared from CUItures of the parent strain and mutant (D-CS,) at early, middle, and late log phase, and the specific activities of the enzymes were established at each phase. From the results in Table 111, it is apparent that the increases in specific activities of alanine racemase and

R E I T Z , S L A D E ,A N D

NEUHAUS

Extract 1. Streptococcus strain Challis 2. Mutant (D-CS,) Sum 3. Combinedb

Alanine Racemase

D-Ala-D-Ala Synthetase

5 . 5 (40).

6.3 (40)

5.6 (5)

9.1 (10)

11.1

15.4

11.2

14.3

The number in parentheses is micrograms of extract protein added. b In 3 the Alanine Racemase Assay contains 40 pg of the parent extract and 5 pg of the mutant extract and the D-Ala-D-Ala Synthetase Assay contains 40 pg of the parent extract and 10 pg of the mutant extract. 5

VOL.

6,

NO.

8,

AUGUST

1967

nous inhibitor or the result of a limitation in an essential cofactor (e.g., pyridoxal phosphate) in the parent strain. In several cases resistance to an antibiotic may be associated with the production of modified enzymes. For example, the elevated dihydrofolate reductase isolated from an amethopterin-resistant Diplococcus pneumoniae (amer-3) has a heat stability that differs from that for the enzyme prepared from the parent strain (Sirotnak et al., 1964). In Figure 1 the heat inactivation of alanine racemase and ~ - A l a - ~ - A l a synthetase is shown. In the presence or absence of ATP, no difference could be detected between the synthetase from the parent strain and that from the mutant strain @-CS,). Moreover, the rate constants for the firstorder decay of alanine racemase were identical for both strains. In addition to the heat stability, the MichaelisMenten constants for the synthetase and racemase have been established for the parent and mutant @-CS,) strain (Figure 2A,B). Identical intercepts on the abscissa (- l/K,) indicate that the Michaelis constants are not different. The inhibition of the synthetase and racemase by D-cycloserine was examined in an attempt to distinguish between the enzymes from mutant @-CS,) and the parent. No difference in the values of Ki could be detected (Figure 3A,B). Thus, this suggests that alanine racemase and D-Ala-D-Ala synthetase of the mutant strain are identical with those of the parent strain. In addition, these results also indicate that formation of an altered enzyme with decreased sensitivity to inhibition by the antibiotic is not a mechanism of resistance in this strain. A mutant strain might be resistant to D-cycloserine if it were capable of forming an altered cell wall in which D-alanine was not an essential component. Cell walls were isolated from the parent and mutant strain @-CS.) and analyzed for amino acids. The results in Figure 4 are presented as molar ratios of amino acid to glutamic acid. These analyses indicate that the walls of the parent and mutant strain @-CSJ’contain almost identical ratios of alanine to glutamic acid. Many of the D-cycloserine-resistant mutants that were isolated did not show increased specific activities of alanine racemase and D-Ala-D-Ala synthetase. For example, mutant (D-CS,) is eightfold resistant to the action of D-cycloserine (Table 11), and yet the levels of the D-cycloserine-sensitive enzymes are not significantly different from those in the parent strain. It seemed likely that the antibiotic resistance of this strain might be the result of an alteration in the permeability of the cells to D-cycloserine. Mora and Snell (1963) reported that Streptococcus faecalis R 8043 contains a transport system for glycine, L-alanine, and D-alanine which requires an energy source, possesses a high temperature coefficient, and is saturated at relatively low external concentrations of alanine. In addition, they reported that D-cycloserine (1 X M) inhibited the accumulation of [14C]alanine. Cells of the parent, mutant (D-CS,), and mutant

RESISTANCE

TO

IO0

80 60

40

RACEMASE (37OC)

t

\

20

5‘

I

I

10

20

MINUTES FIGURE 1: Heat denaturation of alanine racemase and D-Ala-D-Ala synthetase in cell-free extracts of Streptococcus strain Challis and mutant @-CSs). Samples of the extracts were incubated in the following mixtures: (1) “Synthetase ATP (So),” 0.01 M MgCI2;0.01 M ATP neutralized with NaOH; 2 mg of bovine serum albumin/ ml; 0.02 M potassium phosphate buffer (pH 8.0); and 400 pg of extract protein/ml. (2) “Racemase (37”),” all components except ATP and MgC12. (3) “Synthetase (58”),” all components except ATP and MgC12. The samples were incubated at the indicated temperature and aliquots (50 pl) were removed at the indicated time and assayed for enzyme activity according to the procedures described in the Experimental Section. The open symbols represent the parent extract and the closed symbols represent the extract from mutant

+

@-CSJ.

@-CSb) were examined for their ability to concentrate ~-[l~C]alanineand ~-[‘~C]alanine.The results with ~-[l~C]alanine are illustrated in Figure 5A. Mutant @-CS,) and the parent strain concentrate ~-[14C]alanine at almost identical rates. However, mutant @-CS,) lacks the ability to concentrate L-[ 14C]alanine (Figure 5A) and ~-[l~C]alanine (Table V). A preloading procedure similar to that described by Winkler and Wilson (1966) was found to increase the initial rate of uptake of [14C]alanine.The amount accumulated is proportional to time for at least 5 min (Figure 5B). If glucose is omitted or if sodium azide and iodoacetate are added, the uptake is decreased to 10% or less (Figure 5B). After 5 min of accumulation, Mora and Snell (1963) observed that all of the alanine concentrated in S . faecalis R 8043 could be extracted as free alanine. With Streptococcus strain Challis that was not preloaded with alanine, a portion of the radioactivity extracted after 10 min was found in a nucleotide which

D-CYCLOSERINE AND

2565

0 - C A R B A M Y L - D - S E R I N E

BIOCHEMISTRY

I " " ' " "

I

2.0

I

-

( I/[L-ALANINE] ) x IO"

FIGURE 2: Lineweaver-Burk plots of alanine racemase (A) and D-Ala-D-Ala synthetase (B) from Streptococcus strain Challis (0-0) and mutant @-CS,) (0-0).

1 5

IO

15

20

[D-CYCLOSERINE] x IO5

5

IO

15

20

[D-CYCLOSERINE] x IO4

FIGURE 3: Inhibition of D-Ala-D-Ala synthetase (A) and alanine racemase (€3) by D-cycloserine. For A the concentra0 ) 0.005 M, (A, A) 0.01 M. For A and B the open symbols and closed symbols are enzymes tions of D-alanine are (0, from the parent and mutant, respectively.

2566

had an R F equal to that of UDP-NAc-muramylpentapeptide in two solvent systems (Figure 6). The remainder was extracted as free alanine. Thus, in order to minimize secondary reactions, initial rates of accumulation were established during short time intervals (2 min) in preloaded cells (see above). In agreement with the results of Mora and Snell (1963), D-alanine (1 mM) reduces the uptake of L-['~C]alanine (1 mM), and L-alanine (1 mM) reduces the uptake of ~-[l~C]alanine (1 mM) (Table V). This indicates that D-alanine and L-alanine enter Streptococcus strain Challis by the same transport system. In a similar manner, D-cycloserine reduces the uptake of both D[14C]alanineand L-[14C]alanine(Table V).

RFITZ,

SI,ADE,

ANI)

NFIJHAIJS

Resistance to 0-Carbam yl-D-serine. Mutants resistant to 0-carbamyl-D-serine were isolated by the procedure outlined in Table VI. Cell-free extracts were prepared and assayed for alanine racemase, D-Ala-D-Ala synthetase, L-lysine adding enzyme, and D-Ala-D-Ala adding enzyme. As shown in Table VII, an increase in the specific activity of alanine racemase is observed at each step with no significant changes in the specific activities of the other enzymes. Further, a correlation is observed between the minimal inhibitory concentration and the relative specific activity of alanine racemase. Transformation of D-Cycloserine Resistance. The enzyme profiles of five transformants were examined

6,

VOL.

NO.

8,

AUGUST

1967

v: Initial Rates of Accumulation of L-['~C]Alanine and ~-['~C]AIaninein Streptococcus Strain Challis, Mutant (D-CSa), and Mutant (D-CSb):

TABLE

GLUTAMIC ALANINE LYSINE

[14C]AlanineAccumulated (pmoles/min mg-I)

GLUCOSAMINE

~

Additions .

_

I)-[

14C]Alanine 1. Complete system 2. +Sodium azide +Iodoacetate 3. +D-Cycloserine 4. +L-Alanine

Parent

-

D-CS,

D-CSb

MURAMIC

620 48

810

74

-

-

a

GLYCINE ASPARTIC SERINE

E

WENT MUTANT

L-[ 14C]Alanine

1. Complete system 2. +Sodium azide +Iodoacetate 3. +D-Cycloserine 4. fD-Alanine .-

~

-

570 54

720

63

-

-

361 274

-

-

I.o

2 .o

-

0

3.0

RATIO

4: Composition of the cell walls from Streptococcus strain Challis and mutant @-CS,). The bars represent the molar ratio of the compounds to glutamic acid. The amino acid composition was determined by methods described in the Experimental Section. The preparation of walls was described by Struve et a/. (1966). FIGURE

-

-__

the Alanine Accumulation Assay was used with 1 x 10-3 M L- or ~-['~C]alanine, and 180 mg of bacteria (wet weight). The additions were (2) 0.03 M sodium azide and 1 X M iodoacetate; (3) 1 X IOp3 M D-cycloserine; and (4) 1 X lop3M alanine. Blank spaces not determined.

VI: Isolation of 0-Carbamyh-Serine Resistant Mutants.&

TABLE

(Table VIII). These transformants were isolated as described in the Experimental Section from competent cells exposed to DNA isolated from mutant (D-CS,). In contrast to mutant (D-CS,), none of the transformants showed an elevation in both enzymes. However, in every case either D-Ala-D-Ala synthetase or alanine racemase was elevated. The two classes of transformants may be further distinguished by their sensitivity to 0-carbamyh-serine. One class (increased alanine racemase) is tenfold resistant to the action of this antibiotic whereas the other class (increased D-Ala-D-Ala synthetase) is almost as sensitive to this agent as the parent strain (Table VIII).

0-Carbamyl-Dserine* (moles/l. x 103) 0

2 3

4 5 6

7 8 9

Discussion Mutant @-CS,) has elevated levels of the antibioticsensitive enzymes, alanine racemase and D-Ala-D-Ala synthetase. It is not possible with the present results to distinguish between an elevated level (concentration) of enzyme and a modified rate-limiting catalytic constant. In other respects, however, the enzymes from the parent and mutant @-CS,) have identical MichaelisMenten constants, inhibitor constants, and heat stabilities. With regard to mutant @-CS& four alternative mechanisms have been eliminated as possibilities for the acquisition of D-cycloserine resistance. For example, mutant @-CS,) has an alanine-concentrating system

10

13 16 20

Stepe

~_______

1

2

3

Minimal Inhibitory Concn (moles/l. x 103)

>300 6 0 0

2c>300 >300 >300

5

to 3 1c>300 >300 >300 2d

12

20

Approximately lo7 cells were spread on each plate. The colonies growing on the plate with the highest concentration of antibiotic were grown in Todd-Hewitt broth (see mutant isolation) and 10' cells were transferred to the next series. Concentration of O-carbamylD-serine in the blood agar plate. c Mutant grown up for transfer to the next series. d Mutant referred to as mutant (0-CS). e Colonies per plate. 0

2567

BIOCHEMISTRY

12

24

n

P

7

I B

16

‘2 x

X

g 8

4

1

.

I

1

.

1

2

,

3 MINUTES

MINUTES

FIGURE 5 : Accumulation of alanine by Streptococcus strain Challis, mutant (D-CS,), and mutant (D-CSJ. The cells were preloaded and assayed as described in the Alanine Accumulation Assay. In A the following strains were used: (0-0)parent, (0-0) mutant (D-CS,), and (A-A) mutant (D-CSb,).In B the additions are as follows (0-0)complete M iodoacetate. system, (0--0) minus glucose, (A-A) 0.03 M sodium azide and 1 X

Enzyme Profiles in 0-Carbamyl-D-serineResistant Mutanka TABLE VII:

Specific Activity (units/mg of protein) ~~

Enzyme

Parent

Alanine racemase D-Ala-D-Ala synthetase D- Ala-D-Ala adding t-Lysineadding Minimal inhibitory concentration x lo3

~~~

1

2

3h

0.89

3.1

6.6

7.1

0.029

0.030 0.043 0.023

0.45

0.33

0.42

0.40

0.14

0.17

0.15

0.12

1.5

5.0

12

20

.The mutant isolation is described in Table Vi. Mutant (0-CS) described in other experiments is the third-step mutant.

b

which is similar to that of the parent.2 In addition, it does not have the ability to degrade D-cycloserine (unpublished observations). Since the cell walls from the parent and mutant have the same molar ratio of alanine to glutamic acid, there does not appear to be a decreased requirement for alanine. Moreover, the

2568

2 Since D-cycloserine inhibits the accumulation of alanine in S . faecalis (Mora and Snell. 1963) and E. coli W (Kessel and Lubin, 1965), it has been suggested that the concentrating system for alanine also accumulates o-cycloserine.

REITZ, SLADE,

A N D NEUHAUS

inhibitor constants for alanine racemase and D-Ala-DAla synthetase are identical for the enzymes from the parent and mutant (D-CS,). In the mutants selected for resistance to O-carbamylD-serine, only alanine racemase is increased. The production of D-Ala-D-Ala synthetase, D-Ala-D-Ala adding enzyme, and L-lysine adding enzyme is not significantly different from the parent strain. In addition, at each step there is a correlation between the minimal inhibitory concentration and the specific activity of alanine racemase. Therefore, it is concluded that increased production of the drug-sensitive enzymes is a primary mechanism of antibiotic resistance in mutant (D-CS,) and mutant (0-CS). Increased production of the drug-sensitive enzymes has been implicated as a mechanism of resistance in other situations. For example, mutants selected for resistance to 5-methyltryptophan produce elevated levels of the enzymes in the tryptophan pathway (Cohen and Jacob, 1959). Similarly, strains of bacteria resistant to arginine analogs, e.g., canavanine, produce elevated levels of the enzymes of the arginine pathway (Maas, 1961). The elevated levels of alanine racemase and DAla-D-Ala synthetase in mutant (D-CS,) and mutant (0-CS) may be the result of a breakdown in the regulatory mechanism governing the rate of their synthesis. The stepwise development of resistance in the selection of the 0-carbamyh-serine-resistant mutants with the associated increase in enzyme levels may be the cumulative result of mutations which alter the structure of either the repressor or operator (Jacob and Monod, 1961).3 The end result of the selection procedures An alternative explanation for these results would involve changes in the control of constitutive enzyme synthesis.

VOL.

6,

TABLE VIII:

NO.

8,

A U G U S T

1967

Enzyme Profiles in Transformants Resistant to ~-Cycloserine.a Specific Activity (units/mg of protein) Enzyme

Parent

T11

T12

TlA

T2B

Alanine racemase D-Ala-D-Ala synthetase D-Ala-D-Ala adding L-Lysine adding

1.2 0.035 0.35 0.12

6.7 0.033 0.37 0.13

8.9 0.036 0.22 0.09

1.8 0.22 0.60 0.08

1.5 0.21 0.45 b

[D-Cycloserine] X lo4 [O-Carbamyl-~-serine]X l o 3 b

5

2

Minimal Inhibitory Concentration 75 100 50 20 b 5

50 5

a The transformants, T11, T12 were isolated in one series whereas, T1A and T2B were isolated in another series. Not determined.

with 0-carbamyl-D-serine could be a total absence of control in the rate of synthesis of the antibiotic-sensitive enzyme. This is consistent with the numerous isolations of 0-carbamyl-D-serine-resistant mutants in which alanine racemase is increased to the same extent (eightfold). In addition, mutant @-CSJ also possesses an eightfold increase in alanine racemase. In many cases, when the enzymes of a biosynthetic pathway are subject to regulation, all of the enzymes are either coordinately repressed or derepressed. For example, Ames and Garry (1959) observed that the enzymes involved in the biosynthesis of histidine remained in constant proportion to each other over a wide range of repressed and derepressed levels. In contrast, the four enzymes studied in this paper involved in the biosynthesis of UDP-NAc-muramyl-L-Ala-DisoGlu-L-Lys-D-Ala-D-Ala appear to be under independent control. This conclusion is based on the following observations. (1) In mutant @-CSs) alanine racemase and D-Ala-D-Ala synthetase are increased eight- and fivefold, respectively. In contrast, the D-Ala-D-Ah adding enzyme and L-lysine adding enzyme are not increased. (2) In mutant (0-CS), alanine racemase is increased eightfold, whereas the specific activities of the other enzymes are identical with those of the :;arent. (3) Transformation studies show that it is possible to transfer independently the genetic determinants in mutant @-CS,) (increased alanine racemase or increased D-Ala-D-Ala synthetase) to a recipient strain. No transformants were detected with increased levels of both enzymes. Mutant (D-CSb) produces normal levels of the drugsensitive enzymes, alanine racemase and D-Ala-D-Ala synthetase, as well as normal levels of D-Ala-D-Ala adding enzyme and L-lysine adding enzyme. In contrast to mutant @-CS,), this mutant differs in its ability to concentrate alanine. On the basis of the experiments presented in this paper and those presented by Mora and Snell (1963) and Kessel and Lubin (1965), we propose that D-cycloserine is also concentrated by the alanine-transport system. Since mutant @-CSL) pos-

RESISTANCE

sesses a defective transport system for alanine, it would appear that it is also unable to concentrate D-CYC~Oserine. A mutation which results in a decreased permeability

ALANINE

2

4

6 8 DISTANCE (CM)

IO

12

14

FIGURE 6: Chromatography of 14C-labeledcompounds extracted from cells incubated with D-[ 14C]alanineand ~-[I~C]alanine. The hot-water extract was prepared from cells (20 mg dry wt) of Streptococcus strain Challis which had been incubated for 10 min at 25' in a medium that contained 20 mg of glucose, 2 pmoles of [14C]alanine (lo6 cpm), and 50 pmoles of potassium phosphate buffer (pH 8.0) in a volume of 2 ml. The hotwater extract of these cells was chromatographed on Whatman 3MM with 1-butanol-acetic acid-water (200:30:75, v/v) in a descending system. In contrast to the Alanine Accumulation Assay, these cells were not preloaded with alanine. The abbreviations are XAGL, UDP- NAc- muramyl-L- Ala-D- isoGlu-L- Lys; XAG, UDP-NAc-muramyl-L-Ala-D-isoGlu ;XAGLAA, UDPNAc-muramyl-L-Ala-D-isoGlu-L-Lys-D-Ala-D-Ala.

TO D-CYCLOSERINE

AND

2569

0-CARBAMYL-D-SER

BIOCHEMISTRY

to the antibiotic is a common mechanism of drug resistance (Moyed, 1964). For example, Schwartz (1959) isolated two mutants of E. coli that are unable to concentrate D-serine and canavanine, respectively. As a result, they are resistant to the action of these compounds. Recently Unowsky and Rachmeler (1966) observed that E. coli which had acquired antibiotic resistance through the resistance-transfer factor had an altered permeability to several [14C]antibiotics. Curtiss et al. (1965) isolated three cycloserine-resistant mutants (eyer1, cycr2, and cycr3). On the basis of bacterial conjugation experiments, it was concluded that all three mutations were linked to the met1 locus. The mutations in cyc” and cycr3 appear to be in the same gene. They proposed that the mutation in the cycrl strains affects alanine racemase, whereas both the second and third mutations found in cycp3strain affect D-Ala-D-Ala synthetase. It would be very interesting to apply the methods described in this paper to the mutants isolated by Curtiss ef al. (1965). When antibiotics with more than one site of action (e.g., D-cycloserine) are utilized, alterations in permeability may be more important than in those cases where the antibiotic has a single site of action (e.g., 0-carbamyl-D-serine). Decreased uptake of D-cycloserine would “protect” both enzymes, whereas a mutation which caused increased production of one of the sensitive enzymes would still find the biosynthetic sequence restricted by inhibition of the other sensitive enzyme. Except for mutant @-CSa), the other D-CyClOserine-resistant mutants that were isolated resembled mutant @-CSb). On the other hand, all of the mutants which were selected for resistance to 0-carbamyl-Dserine contained increased levels of alanine racemase. Acknowledgments We thank Dr. Robert W. Brockman, Southern Research Institute, for collaboration on the initial phase of this project. The authors also thank Mrs. Marillyn Gragg for amino acid analyses and enzyme assays. We are indebted to Drs. W. G. Struve, R. K. Sinha, and D. Perry, Mr. W. Slamp, and Mrs. Judith Lynch Miller for many discussions and methods which have contributed to this research. References

2570

Ames, B. N., and Garry, B. (1959), Proc. Natl. Acad. Sci. U. S. 45,1453. Benson,R. H. (1966),Anal. Chem. 38,1353. Chambers, P., Bing, J., Lynch, J., Neuhaus, F. C . , and Brockman, R. W. (1963), Bacteriol. Proc. 119. Ciak, J., and Hahn, F. E. (1959), Antibiot. Chemotherapy 9,47. Cohen, G., and Jacob, F. (1959), Compt. Rend. 248, 3490. Curtiss, R., Charamella, L. J., Berg, C. M., and Harris, P. E. (1969, J. Bacteriol. 90,1238. Howe, W. B., Melson, H. L., Meridith, C. H., Morrison, J. R., Platt, M. H., and Strominger, J. L. (1964),

REITZ,

SLADE,

A N D

NEUHAUS

J. Pharmacol. 143,282. Ito, E., and Strominger, J. L. (1962a), J. Biol. Chem. 237,2689. Ito, E., and Strominger, J. L. (1962b), J. Biol. Chem. 239,210 Jacob, F., and Monod, J. (1961), J. Mol. Biol.3,318. Kessel, D., and Lubin, M. (1965), Biochemistry 4, 561. Lowry, 0. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. (1951), J . Biol. Chem. 193,265. Lynch, J. L., and Neuhaus, F. C. (1966), J. Bacteriol. 91,449. Maas, W. K. (1961), Cold Spring Hurbor Sjwp. Quanr. Biol. 26,183. Mora, F., and Snell, E. E. (1963), Biochemistry 2,163. Moyed, H. S. (1964), Ann. Rev. Microbiol. 18,347. Neuhaus,F. C . (1962), J . Biol. Chem. 237,778. Neuhaus, F. C. (1967), in Mechanism of Action and Biosynthesis of Antibiotics, Vol. 1, Gottlieb, D., and Shaw, P. L., Ed., Heidelberg, Springer-Verlag, p 40. Neuhaus, F. C., and Lynch, J. L. (1964), Biochemistry 3,471. Neuhaus, F. C., and Struve, W. G . (1965), Biochemistry 4,120. Park, J. T. (1958), Biochem. J. 70,2P. Park, J. T. (1960), Antimicrobial Agents Ann., 338. Patterson, M. S., and Greene, R. C. (1965), Anal. Chem. 37,854. Perry, D., and Slade, H. D. (1962),J. Bacteriol. 83,443. Perry, D., and Slade, H. D. (1966), J . Bacteriol. 91,2216. Reitz, R. H. (1966), Ph.D. Thesis, Northwestern University, Evanston, Ill. Reitz, R. H., Slade, H. D., and Neuhaus, F. C. (1966), Federation Proc. 25,344. Schwartz, J. H. (1959), Biochim. Biophys. Acta 32,582. Shockman, G. D. (1959), Proc. SOC.Exptl. Biol. Med. 101,693. Sirotnak, F. M., Donati, G. J., and Hutchison, D. J. (1964), J . Biol. Chem. 239,4298. Spackman, D. H., Stein, W. H., and Moore, S. (1958), Anal. Chem. 30,1190. Stickgold, R. A,, and Neuhaus, F. C. (1967), J . Biol. Chem. 242,1331. Strominger, J. L. (1962), Federation Proc. 21,134. Strominger, J. L., Ito, E., Threnn, R. H. (1960), J Am. Chem. SOC. 82,998. Strominger, J. L., and Threnn, R. H. (1959), Biochim. Biophys. Acta 36,83. Strominger, J . L., Threnn, R. H., and Scott, S. S. (1959), J. Am. Chem. SOC.81,3803. Struve, W. G., Sinha, R. K., and Neuhaus, F. C . (1966), Biochemistry 5,82. Swift, H. F., Wilson, A . T., and Lancefield, R . C. (1943), J . Exptl. Med. 78,127. Tanaka, N., Sashikata, K., Wada, T., Sugawara, S . , and Umezawa, H. (1963), J. Antibiotics (Tokyo) 16, 217. Unowsky, J , and Rachmeler, M. (19661, J. BacterioL. 92,358. Winkler, H. H., and Wilson, T. H. (1966), J. Biol. Chem. 241,2200.